Matching donations without crowding out? : some theoretical considerations and a field experiment

Is there a way of matching donations that avoids crowding out? And, more generally, what is the best way to utilize a bigger lead gift for raising smaller contribu-tions in a fundraising campaign? To answer these questions, we present a novel matching method, some simple theoretical considerations, and evidence from a large-scale natural field experiment on charitable giving. We compare a standard linear matching scheme with a novel matching scheme in which the matched amount is allocated towards a different project. Treatments with unconditional lead gifts serve as controls. Similar to findings from the previous literature, conventional matching for the same project results in partial crowding-out (increasing, however, the participation rate and the overall return per mail-out). The novel matching scheme that we propose also increases the response rate and, in addition, does avoid crowding out.